gem5::AMDGPU::fp16_e5m10_info Union Reference

#include <fp16_e5m10.hh>

Public Types
enum	bitSizes {   ebits = 5 , mbits = 10 , sbits = 1 , zbits = 16 ,   bias = 15 , inf = 0x7c000000 , nan = 0x7c100000 , max = 0x7bff0000 }

Public Attributes
uint32_t	storage
struct {
unsigned   zero: zbits
unsigned   mant: mbits
unsigned   exp: ebits
unsigned   sign: sbits
};

Detailed Description

Definition at line 43 of file fp16_e5m10.hh.

Member Enumeration Documentation

bitSizes

enum gem5::AMDGPU::fp16_e5m10_info::bitSizes

Enumerator
ebits
mbits
sbits
zbits
bias
inf
nan
max

Definition at line 45 of file fp16_e5m10.hh.

Member Data Documentation

[struct]

struct { ... } gem5::AMDGPU::fp16_e5m10_info

exp

unsigned gem5::AMDGPU::fp16_e5m10_info::exp

Definition at line 63 of file fp16_e5m10.hh.

mant

unsigned gem5::AMDGPU::fp16_e5m10_info::mant

Definition at line 62 of file fp16_e5m10.hh.

sign

unsigned gem5::AMDGPU::fp16_e5m10_info::sign

Definition at line 64 of file fp16_e5m10.hh.

storage

uint32_t gem5::AMDGPU::fp16_e5m10_info::storage

Definition at line 58 of file fp16_e5m10.hh.

Referenced by std::numeric_limits< gem5::AMDGPU::fp16_e5m10_info >::infinity(), std::numeric_limits< gem5::AMDGPU::fp16_e5m10_info >::max(), and std::numeric_limits< gem5::AMDGPU::fp16_e5m10_info >::quiet_NaN().

zero

unsigned gem5::AMDGPU::fp16_e5m10_info::zero

Definition at line 61 of file fp16_e5m10.hh.

---

The documentation for this union was generated from the following file:

• arch/amdgpu/common/dtype/fp16_e5m10.hh